Covariance Matrix Estimation Errors and Diagonal Loading in Adaptive Arrays

摘要

The use of adaptive beamforming techniques in which the radiation pattern of arrays of elements may be adaptively changed to reject interference signals can provide significant performance advantages in radar, communication, and sonar applications. Simulations have been used to investigate the effect of covariance matrix sample matrix inversion algorithm. Inadequate estimation of the covariance matrix results in adapted antenna patterns with high sidelobes and distorted main beams. A technique to reduce these effects by modifying the covariance matrix estimate is described from the point of view of eigenvector decomposition. This technique reduces the system nulling capability against low-level interference, but parametric studies show that it is an effective approach in many situations. Keywords: Reprints. (kr)